Run 4. Outer Step 21. Inner Step 21.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5526171,
-    5526177,
-    5526183,
-    5526189,
-    5526195
+    5526202,
+    5526208,
+    5526214,
+    5526220,
+    5526227
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 20,
+  "inner_step": 21,
   "inner_steps": 0,
   "last_allreduce_block": 5523943,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:03eef167647f856b403352b15029314220721af542d89677933913c7dd105fdc
+oid sha256:b30766ae0b2a851713968e029cf8cbcdcedd9d21639e8eeb62a2fab914639038
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dafd9cca8482fda7020c0b69bb29f10550fb79dda5a0ce857b121ff6e3728b84
+oid sha256:49dbc786145967e8e06c958a2152003841d5ba5921c82503bdd374d3ae4dc619
 size 4040701744